Biallelic human SHARPIN loss of function induces autoinflammation and immunodeficiency.

The linear ubiquitin assembly complex (LUBAC) consists of HOIP, HOIL-1 and SHARPIN and is essential for proper immune responses. Individuals with HOIP and HOIL-1 deficiencies present with severe immunodeficiency, autoinflammation and glycogen storage disease. In mice, the loss of Sharpin leads to severe dermatitis due to excessive keratinocyte cell death. Here, we report two individuals with SHARPIN deficiency who manifest autoinflammatory symptoms but unexpectedly no dermatological problems. Fibroblasts and B cells from these individuals showed attenuated canonical NF-κB responses and a propensity for cell death mediated by TNF superfamily members. Both SHARPIN-deficient and HOIP-deficient individuals showed a substantial reduction of secondary lymphoid germinal center B cell development. Treatment of one SHARPIN-deficient individual with anti-TNF therapies led to complete clinical and transcriptomic resolution of autoinflammation. These findings underscore the critical function of the LUBAC as a gatekeeper for cell death-mediated immune dysregulation in humans.

[Clinical phenotype and genetic characteristics of a Chinese pedigree affected with Spastic paraplegia type 5A].

OBJECTIVE: To explore the clinical phenotype and genetic characteristics of a Chinese pedigree affected with Spastic paraplegia type 5A (SPG5A). METHODS: A pedigree suspected for Hereditary spastic paraplegia (HSP) at Henan Children's Hospital on August 15 2022 was selected as the study subject. Clinical data of the pedigree was collected. Peripheral blood samples were collected from members of the pedigree. Following extraction of genomic DNA, trio-WGS was carried out, and candidate variant was verified by Sanger sequencing. RESULTS: The child, a 1-year-old boy, had presented with microcephaly, hairy face and dorsal side of distal extremities and trunk, intellectual and motor development delay, increased muscle tone of lower limbs, hyperreflexes of bilateral knee tendons, and positive pathological signs. His parents and sister both had normal phenotypes. Trio-WGS revealed that the child has harbored a homozygous c.1250G>A (p.Arg417His) variant of the CYP7B1 gene, for which his mother was heterozygous, the father and sister were of the wild type. The variant was determined to have originated from maternal uniparental disomy (UPD). The result of Sanger sequencing was in keeping with the that of trio-WGS. SPG5A due to maternal UPD of chromosome 8 was unreported previously. CONCLUSION: The child was diagnosed with SPG5A, a complex type of HSP, for which the homozygous c.1250G>A variant of the CYP7B1 gene derived from maternal UPD may be accountable.

HD-Zip proteins modify floral structures for self-pollination in tomato.

Cleistogamy is a type of self-pollination that relies on the formation of a stigma-enclosing floral structure. We identify three homeodomain-leucine zipper IV (HD-Zip IV) genes that coordinately promote the formation of interlocking trichomes at the anther margin to unite neighboring anthers, generating a closed anther cone and cleistogamy (flower morphology necessitating strict self-pollination). These HD-Zip IV genes also control style length by regulating the transition from cell division to endoreduplication. The expression of these HD-Zip IV genes and their downstream gene, Style 2.1, was sequentially modified to shape the cleistogamy morphology during tomato evolution and domestication. Our results provide insights into the molecular basis of cleistogamy in modern tomato and suggest targets for improving fruit set and preventing pollen contamination in genetically modified crops.

Deep Self-Reconstruction Fusion Similarity Hashing for the Diagnosis of Alzheimer's Disease on Multi-Modal Data.

The pathogenesis of Alzheimer's disease (AD) is extremely intricate, which makes AD patients almost incurable. Recent studies have demonstrated that analyzing multi-modal data can offer a comprehensive perspective on the different stages of AD progression, which is beneficial for early diagnosis of AD. In this paper, we propose a deep self-reconstruction fusion similarity hashing (DS-FSH) method to effectively capture the AD-related biomarkers from the multi-modal data and leverage them to diagnose AD. Given that most existing methods ignore the topological structure of the data, a deep self-reconstruction model based on random walk graph regularization is designed to reconstruct the multi-modal data, thereby learning the nonlinear relationship between samples. Additionally, a fused similarity hash based on anchor graph is proposed to generate discriminative binary hash codes for multi-modal reconstructed data. This allows sample fused similarity to be effectively modeled by a fusion similarity matrix based on anchor graph while modal correlation can be approximated by Hamming distance. Especially, extracted features from the multi-modal data are classified using deep sparse autoencoders classifier. Finally, experiments conduct on the AD Neuroimaging Initiative database show that DS-FSH outperforms comparable methods of AD classification. To conclude, DS-FSH identifies multi-modal features closely associated with AD, which are expected to contribute significantly to understanding of the pathogenesis of AD.

Molecular Imaging of Alzheimer's Disease-Related Sigma-1 Receptor in the Brain via a Novel Ru-Mediated Aromatic 18F-deoxyfluorination Probe.

Sigma-1 receptor (σ1R) is an intracellular protein implicated in a spectrum of neurodegenerative conditions, notably Alzheimer's disease (AD). Positron emission tomography (PET) imaging of brain σ1R could provide a powerful tool for better understanding the underlying pathomechanism of σ1R in AD. In this study, we successfully developed a 18F-labeled σ1R radiotracer [18F]CNY-05 via an innovative ruthenium (Ru)-mediated 18F-deoxyfluorination method. [18F]CNY-05 exhibited preferable brain uptake, high specific binding, and slightly reversible pharmacokinetics within the PET scanning time window. PET imaging of [18F]CNY-05 in nonhuman primates (NHP) indicated brain permeability, metabolic stability, and safety. Moreover, autoradiography and PET studies of [18F]CNY-05 in the AD mouse model found a significantly decreased brain uptake compared to that in wild-type mice. Collectively, we have provided a novel 18F-radiolabeled σ1R PET probe, which enables visualizing brain σ1R in health and neurological diseases.

Characterization of the complete chloroplast genome sequence of Camellia tetracocca (Theaceae).

Camellia tetracocca H.T. Chang 1981 is an important wild relative of cultivated tea plants. Its leaves are widely used by local people to make tea, showing great economic and breeding values. We here report the complete chloroplast genome of C. tetracocca using Illumina sequencing technology. The complete chloroplast genome of C. tetracocca is 157,026 bp in length, and structurally contains a pair of inverted repeat regions (IRa and IRb, 26,052 bp) separated by a large single-copy region (LSC, 86,669 bp) and a small single-copy region (SSC, 18,253 bp). It is composed of 131 predicted genes, including 86 protein-coding genes, 37 transfer RNA genes, and eight ribosomal RNA genes. The overall GC content is 37.31%. Phylogenetic analysis among four Camellia species and 11 other close species reveals a close relationship between C. tetracocca and C. gymnogyna.

Reduced contrast sensitivity function is correlated with changes to cone photoreceptors in simple high myopia.

Purpose: To investigate the contrast sensitivity function (CSF) changes in simple high myopia (SHM) and evaluate the correlations between these changes with the early changes in the retinal microstructure. Methods: This prospective study comprised 81 subjects, 20 with emmetropia (EM), 26 with low myopia and moderate myopia (LM/MM), and 35 with SHM. The area under the log CSF curve (AULCSF) and the cut-off spatial frequency (Cut-off SF) were employed as measures of CSF. Adaptive optics (AO) was employed to quantify the cone density, spacing, and regularity. The thickness and blood flow of the retinal sublayers were determined from vertical and horizontal optical coherence tomography angiography (OCTA) A-scans. Swept-source optical coherence tomography (SS-OCT) was employed to analyze the choroidal thickness (CT) and choroidal vascularity using a custom algorithm. Differences in the retinal and choroidal parameters, cone distribution, AULCSF, and Cut-off SF were compared among the three groups. Multivariate linear mixed models were used to elucidate the associations between photoreceptor morphological alterations, retinal and choroidal parameters, and AULCSF. Results: The AULCSF and Cut-off SF were significantly lower in the SHM group compared to the EM and LM groups (p < 0.05). The SHM group had less cone density, larger cone spacing, and lower cone regularity than the EM and LM/MM groups (p < 0.05). Moreover, the thickness of the inner segment of photoreceptors (IS), retinal pigment epithelium (RPE) layer and choroid were reduced, and the outer segment of photoreceptors (OS) was thicker in the SHM group compared to the EM and LM/MM groups (all p < 0.05). A longer axial length (AL) was correlated with decreased AULCSF, cone density, and cone spacing (r = -0.800 to 0.752, all p < 0.050). Additionally, decreased CSF was correlated with lower cone density (r = 0.338, p = 0.035). Conclusion: Decreased contrast sensitivity was observed in patients with SHM and cone density was significantly correlated with reduced AUCSF.

Integrative transcriptomic and metabolomic analyses reveals the toxicity and mechanistic insights of bioformulated chitosan nanoparticles against Magnaporthe oryzae.

Rice blast, an extremely destructive disease caused by the filamentous fungal pathogen Magnaporthe oryzae, poses a global threat to the production of rice (Oryza sativa L.). The emerging trend of reducing dependence on chemical fungicides for crop protection has increased interest in exploring bioformulated nanomaterials as a sustainable alternative antimicrobial strategy for effectively managing plant diseases. Herein, we used physiomorphological, transcriptomic, and metabolomic methods to investigate the toxicity and molecular action mechanisms of moringa-chitosan nanoparticles (M-CNPs) against M. oryzae. Our results demonstrate that M-CNPs exhibit direct antifungal properties by impeding the growth and conidia formation of M. oryzae in a concentration-dependent manner. Propidium iodide staining indicated concentration-dependent significant apoptosis (91.33%) in the fungus. Ultrastructural observations revealed complete structural damage in fungal cells treated with 200 mg/L M-CNPs, including disruption of the cell wall and destruction of internal organelles. Transcriptomic and metabolomic analyses revealed the intricate mechanism underlying the toxicity of M-CNPs against M. oryzae. The transcriptomics data indicated that exposure to M-CNPs disrupted various processes integral to cell membrane biosynthesis, aflatoxin biosynthesis, transcriptional regulation, and nuclear integrity in M. oryzae., emphasizing the interaction between M-CNPs and fungal cells. Similarly, metabolomic profiling demonstrated that exposure to M-CNPs significantly altered the levels of several key metabolites involved in the integral components of metabolic pathways, microbial metabolism, histidine metabolism, citrate cycle, and lipid and protein metabolism in M. oryzae. Overall, these findings demonstrated the potent antifungal action of M-CNPs, with a remarkable impact at the physiological and molecular level, culminating in substantial apoptotic-like fungal cell death. This research provides a novel perspective on investigating bioformulated nanomaterials as antifungal agents for plant disease control.

Application of dual chemotherapeutic drug delivery system based on metal-organic framework platform in enhancing tumor regression for breast cancer research.

The nanomedicine system based on dual drug delivery systems (DDDs) can significantly enhance the efficacy of tumor treatment. Herein, a metal-organic framework, Zeolite imidazole salt frames 8 (ZIF-8), was successfully utilized as a carrier to load the dual chemotherapeutic drugs doxorubicin (DOX) and camptothecin (CPT), named DOX/CPT@ZIF-8 (denoted as DCZ), and their inhibitory effects on 4T1 breast cancer cells were evaluated. The study experimentally demonstrated the synergistic effects of the dual chemotherapeutic drugs within the ZIF-8 carrier and showed that the ZIF-8 nano-carrier loaded with the dual drugs exhibited stronger cytotoxicity and inhibitory effects on 4T1 breast cancer cells compared to single-drug treatment. The use of a ZIF-8-based dual chemotherapeutic drug carrier system highlighted its potential advantages in suppressing 4T1 breast cancer cells.

Pharmacokinetics and safety of a new generic lurasidone: a phase I bioequivalence study in healthy Chinese subjects.

Latuda® is a novel antipsychotic drug for schizophrenia and bipolar depression. A bioequivalence trial was performed to investigate the bioequivalence of Latuda® and its generic drug lurasidone. Two independent trials were carried out, each involving 28 subjects. In the fasting trial, subjects were randomly assigned to two groups (1:1 ratio), receiving either 40 mg of generic lurasidone or Latuda®. After a 7-day washout period, subjects entered the second period with a crossover administration of 40 mg of generic lurasidone or Latuda®. The postprandial study design was similar to that of the fasting study. In the fasting study, the pharmacokinetic (PK) parameter values of generic lurasidone and Latuda® were as follows: the Cmax was 28.84 ± 19.34 ng/ml and 28.22 ± 21.19 ng/ml, respectively; the AUC0-t was 121.39 ± 58.47 h*ng/ml and 118.35 ± 52.24 h*ng/ml, respectively; and the AUC0-∞ was 129.63 ± 63.26 h*ng/ml and 126.59 ± 57.99 h*ng/ml, respectively. The primary pharmacokinetic parameter, Cmax, was assessed for equivalence using reference-scaled average bioequivalence (RSABE), while other parameters (AUC0-t, AUC0-∞) were evaluated using average bioequivalence (ABE). The results indicate that both Cmax and AUC meet the equivalence criteria. In the postprandial study, the PK values of generic lurasidone and Latuda® were as follows: the Cmax was 74.89 ± 32.06 ng/ml and 83.51 ± 33.52 ng/ml, respectively; the AUC0-t was 274.77 ± 103.05 h*ng/ml and 289.26 ± 95.25 h*ng/ml, respectively; and the AUC0-∞ was 302.44 ± 121.60 h*ng/ml and 316.32 ± 109.04 h*ng/ml, respectively. The primary pharmacokinetic parameters (Cmax, AUC0-t, AUC0-∞) were assessed for equivalence using ABE, and both met the equivalence criteria. In the study, lurasidone and Latuda® both exhibited acceptable safety and tolerability. The results displayed that lurasidone and Latuda® were bioequivalent and safe in healthy Chinese participants. Clinical Trial Registry: This trial is registered at chinadrugtrials.org.cn (no.: CTR20191717, date: 2019.08.29).

Milk: A Natural Guardian for the Gut Barrier.

The gut barrier plays an important role in health maintenance by preventing the invasion of dietary pathogens and toxins. Disruption of the gut barrier can cause severe intestinal inflammation. As a natural source, milk is enriched with many active constituents that contribute to numerous beneficial functions, including immune regulation. These components collectively serve as a shield for the gut barrier, protecting against various threats such as biological, chemical, mechanical, and immunological threats. This comprehensive review delves into the active ingredients in milk, encompassing casein, α-lactalbumin, ß-lactoglobulin, lactoferrin, the milk fat globular membrane, lactose, transforming growth factor, and glycopeptides. The primary focus is to elucidate their impact on the integrity and function of the gut barrier. Furthermore, the implications of different processing methods of dairy products on the gut barrier protection are discussed. In conclusion, this study aimed to underscore the vital role of milk and dairy products in sustaining gut barrier health, potentially contributing to broader perspectives in nutritional sciences and public health.

Diffusion models in bioinformatics and computational biology.

Denoising diffusion models embody a type of generative artificial intelligence that can be applied in computer vision, natural language processing and bioinformatics. In this Review, we introduce the key concepts and theoretical foundations of three diffusion modelling frameworks (denoising diffusion probabilistic models, noise-conditioned scoring networks and score stochastic differential equations). We then explore their applications in bioinformatics and computational biology, including protein design and generation, drug and small-molecule design, protein-ligand interaction modelling, cryo-electron microscopy image data analysis and single-cell data analysis. Finally, we highlight open-source diffusion model tools and consider the future applications of diffusion models in bioinformatics.

LET-767 determines lipid droplet protein targeting and lipid homeostasis.

Lipid droplets (LDs) are composed of a core of neutral lipids wrapped by a phospholipid (PL) monolayer containing several hundred proteins that vary between different cells or organisms. How LD proteins target to LDs is still largely unknown. Here, we show that RNAi knockdown or gene mutation of let-767, encoding a member of hydroxysteroid dehydrogenase (HSD), displaced the LD localization of three well-known LD proteins: DHS-3 (dehydrogenase/reductase), PLIN-1 (perilipin), and DGAT-2 (diacylglycerol O-acyltransferase 2), and also prevented LD growth in Caenorhabditis elegans. LET-767 interacts with ARF-1 (ADP-ribosylation factor 1) to prevent ARF-1 LD translocation for appropriate LD protein targeting and lipid homeostasis. Deficiency of LET-767 leads to the release of ARF-1, which further recruits and promotes translocation of ATGL-1 (adipose triglyceride lipase) to LDs for lipolysis. The displacement of LD proteins caused by LET-767 deficiency could be reversed by inhibition of either ARF-1 or ATGL-1. Our work uncovers a unique LET-767 for determining LD protein targeting and maintaining lipid homeostasis.

Bio-formulated chitosan nanoparticles enhance disease resistance against rice blast by physiomorphic, transcriptional, and microbiome modulation of rice (Oryza sativa L.).

Rice blast disease (RBD) caused by Magnaporthe oryzae, threaten food security by cutting agricultural output. Nano agrochemicals are now perceived as sustainable, cost-effective alternatives to traditional pesticides. This study investigated bioformulation of moringa chitosan nanoparticles (M-CsNPs) and their mechanisms for suppressing RBD while minimizing toxic effects on the microenvironment. M-CsNPs, sized 46 nm with semi-spherical morphology, significantly suppressed pathogen growth, integrity, and colonization at 200 mg L-1in vitro. Greenhouse tests with foliar exposure to the same concentration resulted in a substantial 77.7 % reduction in RBD, enhancing antioxidant enzyme activity and plant health. Furthermore, M-CsNPs improved photosynthesis, gas exchange, and the nutritional profile of diseased rice plants. RNA-seq analysis highlighted upregulated defense-related genes in treated rice plants. Metagenomic study showcased reshaping of the rice microbiome, reducing Magnaporthe abundance by 93.5 %. Both healthy and diseased rice plants showed increased microbial diversity, particularly favoring specific beneficial species Thiobacillus, Nitrospira, Nocardioides, and Sphingomicrobium in the rhizosphere and Azonexus, Agarivorans, and Bradyrhizobium in the phyllosphere. This comprehensive study unravels the diverse mechanisms by which M-CsNPs interact with plants and pathogens, curbing M. oryzae damage, promoting plant growth, and modulating the rice microbiome. It underscores the significant potential for effective plant disease management.

Enhancing the anti-thixotropic properties of waxy maize starch modified by different α-amylases and its underlying molecular mechanism.

The large thixotropy of the starch-thickened foods is often unfavorable in many applications. This study examined the contribution of the proportion of amylopectin chain length to time-dependence of starch gels. The α-amylase (AM) from Bacillus stearothermophilus and maltogenic α-amylase (MA) from Bacillus subtilis were used to trim amylopectin in different reaction patterns. HPLC, HPAEC and IBC data suggested AM attacked B-chains (DP 12-36), causing an increment in number of the chains with DP 6-12, whereas MA primarily trimmed the short B-chains (DP 12-18) and partial A-chains (DP 9-12) to generate short chains with DP 6-9. Interestingly, the recovery of AM-gels was faster than MA-gels at the same degree of hydrolysis when subjected to shear according to the linear correlation analysis. When releasing the same mass of sugar, shortening of the long internal chains played an important role in reducing time dependence of starch gel rather than the external side chains. Possible models were proposed to illustrate the differences in the mechanism of rapid-recovery caused by different side-chain distributions. The outcome provided a new perspective to regulate the thixotropy behavior of starch through enzyme strategies in the granular state.

High performance polyvinyl alcohol/lignin fibers with excellent mechanical and water resistance properties.

To address the shortcoming of Polyvinyl alcohol (PVA) fibers for food or medical packaging materials including low mechanical strength and poor water resistance, lignin (LN) was used as raw material, acetone/H2O as solvent to self-assemble into lignin nanoparticles (LNP) by adverse solvent precipitation approach, and then PVA/LNP composite fibers with different LNP contents were fabricated successfully by wet and dry spinning. Herein, vast hydrophilic hydroxyl groups in PVA decreased owing to the hydrogen bond between LN and PVA, Especially, with only 0.5 wt% loading of LNP into the PVA/LNP fibers, the diameter was 94.4 dtex, tensile strength was 10.1 cN/dtex (1279.8 MPa), initial modulus was 94.7 cN/dtex (12.0 GPa), the crystallinity was 56.7 %, the orientation was 97.1 %, and water contact angle was 103.1°. Compared with pure PVA fibers, the tensile strength of PVA/LNP-0.5 fibers was increased by 44.2 % and the contact angle was increased 37°. This work provides novel insights into obtaining lignin-reinforced PVA composite fibers with strong mechanical properties and excellent water resistance properties, indicating the potential of the PVA/LNP fibers for food or medical packaging application.

Insights into the Structure-Function Relationship of GH70 GtfB α-Glucanotransferases from the Crystal Structure and Molecular Dynamic Simulation of a Newly Characterized Limosilactobacillus reuteri N1 GtfB Enzyme.

α-Glucanotransferases of the CAZy family GH70 convert starch-derived donors to industrially important α-glucans. Here, we describe characteristics of a novel GtfB-type 4,6-α-glucanotransferase of high enzyme activity (60.8 U mg-1) from Limosilactobacillus reuteri N1 (LrN1 GtfB), which produces surprisingly large quantities of soluble protein in heterologous expression (173 mg pure protein per L of culture) and synthesizes the reuteran-like α-glucan with (α1 → 6) linkages in linear chains and branch points. Protein structural analysis of LrN1 GtfB revealed the potential crucial residues at subsites -2∼+2, particularly H265, Y214, and R302, in the active center as well as previously unidentified surface binding sites. Furthermore, molecular dynamic simulations have provided unprecedented insights into linkage specificity hallmarks of the enzyme. Therefore, LrN1 GtfB represents a potent enzymatic tool for starch conversion, and this study promotes our knowledge on the structure-function relationship of GH70 GtfB α-glucanotransferases, which might facilitate the production of tailored α-glucans by enzyme engineering in future.

A gradient of the HD-Zip regulator Woolly regulates multicellular trichome morphogenesis in tomato.

Homeodomain (HD) proteins regulate embryogenesis in animals such as the fruit fly (Drosophila melanogaster), often in a concentration-dependent manner. HD-leucine zipper (Zip) IV family genes are unique to plants and often function in the L1 epidermal cell layer. However, our understanding of the roles of HD-Zip IV family genes in plant morphogenesis is limited. In this study, we investigated the morphogenesis of tomato (Solanum lycopersicum) multicellular trichomes, a type of micro-organ in plants. We found that a gradient of the HD-Zip IV regulator Woolly (Wo) coordinates spatially polarized cell division and cell expansion in multicellular trichomes. Moreover, we identified a TEOSINTE BRANCHED1, CYCLOIDEA, and PROLIFERATING CELL NUCLEAR ANTIGEN BINDING FACTOR (TCP) transcription factor-encoding gene, SlBRANCHED2a (SlBRC2a), as a key downstream target of Wo that regulates the transition from cell division to cell expansion. High levels of Wo promote cell division in apical trichome cells, whereas in basal trichome cells, Wo mediates a negative feedback loop with SlBRC2a that forces basal cells to enter endoreduplication. The restricted high and low activities of Wo patterns the morphogenesis of tomato multicellular trichomes. These findings provide insights into the functions of HD-Zip IV genes during plant morphogenesis.

Development and Characterization of a Novel Carbon-11 Labeled Positron Emission Tomography Radiotracer for Neuroimaging of Sirtuin 1 with Benzoxazine-Based Compounds.

Introduction: Sirtuins (SIRTs) comprise a group of histone deacetylase enzymes crucial for regulating metabolic pathways and contributing significantly to various disease mechanisms. Sirtuin 1 (SIRT1), among the seven known mammalian homologs, is extensively investigated and understood, playing a key role in neurodegenerative disorders and cancer. This study focuses on potential as a therapeutic target for conditions such as Parkinson's disease (PD), Huntington's disease (HD), and Alzheimer's disease (AD). Methods: Utilizing positron emission tomography (PET) as a noninvasive molecular imaging modality, we aimed to expedite the validation of a promising sirtuin 1 inhibitor for clinical trials. However, the absence of a validated sirtuin 1 PET radiotracer impedes clinical translation. We present the development of [11C]1, and 11C-labeled benzoxazine-based derivative, as a lead imaging probe. The radiosynthesis of [11C]1 resulted in a radiochemical yield of 31 ± 4%. Results: Baseline studies demonstrated that [11C]1 exhibited excellent blood-brain barrier (BBB) penetration capability, with uniform accumulation throughout various brain regions. Self-blocking studies revealed that introducing an unlabeled compound 1, effectively blocking sirtuin 1, led to a substantial reduction in whole-brain uptake, emphasizing the in vivo specificity of [11C]1 for sirtuin 1. Discussion: The development of [11C]1 provides a valuable tool for noninvasive imaging investigations in rodent models with aberrant sirtuin 1 expression. This novel radiotracer holds promise for advancing our understanding of sirtuin 1's role in disease mechanisms and may facilitate the validation of sirtuin 1 inhibitors in clinical trials.

Benchmarking Supervised and Self-Supervised Learning Methods in A Large Ultrasound Muti-task Images Dataset.

Deep learning in ultrasound(US) imaging aims to construct foundational models that accurately reflect the modality's unique characteristics. Nevertheless, the limited datasets and narrow task types have restricted this field in recent years. To address these challenges, we introduce US-MTD120K, a multi-task ultrasound dataset with 120,354 real-world two-dimensional images. This dataset covers three standard plane recognition and two diagnostic tasks in ultrasound imaging, providing a rich basis for model training and evaluation. We detail the data collection, distribution, and labelling processes, ensuring a thorough understanding of the dataset's structure. Furthermore, we conduct extensive benchmark tests on 27 state-of-the-art methods from both supervised and self-supervised learning(SSL) perspectives. In the realm of supervised learning, we analyze the sensitivity of two main feature computation methods to ultrasound images at the representational level, highlighting that models which judiciously constrain global feature computation could potentially serve as a viable analytical approach for US image analysis. In the context of self-supervised learning, we delved into the modelling process of self-supervised learning models for medical images and proposed an improvement strategy, named MoCo-US, a solution that addresses the excessive reliance on pretext task design from the input side. It achieves competitive performance with minimal pretext task design and enhances other SSL methods simply. The dataset and the code will be available at https://github.com/JsongZhang/CDOA-for-UMTD.